<div dir="ltr"><div>Kitware and the VTK team are happy to announce that VTK 6.1 has entered the release candidate stage!</div><div><br></div><div>You can find the source, data, and new vtkpython binary packages here:</div>
<div><a href="http://www.vtk.org/VTK/resources/software.html#cand">http://www.vtk.org/VTK/resources/software.html#cand</a></div><div><br></div><div>Please try this version of VTK and report any issues to the list or the bug tracker so that we can try to address them before VTK 6.1.0 final.</div>
<div><br></div><div>The official release notes will be available when VTK final is released in the next few weeks. In the meantime here is a preview:</div><div><br></div><div>VTK continues to evolve to make use of the parallelism available in modern architectures. In this release we introduce the vtkDax accelerator module and vtkSMP multithreaded filters. The Dax interface allows VTK to offload work to Sandia's Dax framework for GPGPU and multi-core data processing. Inria's vtkSMP brings a framework for fine-grained work-stealing parallel processing inside of unstructured data processing filters.</div>
<div><br></div><div>In this release VTK also takes a giant leap onto the web processing frontier. ParaView's core web interface and widget components have been promoted to VTK. Several web applications are provided as examples that show how to use VTK processing and rendering in a Web context. VTK's web interface also includes newly developed features such as an integrated, object-oriented python framework for writing dashboard tests of web applications, a simple python-based launcher and file upload server, and a new viewport that drives the VGL (webgl-based) rendering library developed by Kitware and NYU.</div>
<div><br></div><div>Another major advance is the introduction of the vtkMappedDataArray class and a related set of changes that create a new abstract interface to external data storage. This allows, for example, zero copy import from and export to external simulation packages which may layout their data differently than VTK does.</div>
<div><br></div><div>There are numerous improvements to VTK's low level OpenGL interfaces. These include better error reporting, allowing depth peeled translucency on AMD graphics chipssets, and allowing the use of advanced rendering features available in modern OSMesa releases. Concurrently we've also removed the dated and little used XML and Cg shaders.</div>
<div><br></div><div>We've made many improvements to VTK's build system in this release. We now use private dependencies among VTK modules to avoid exporting implementation symbols and shorten the link line length. To simplify the creation of applications that depend on subsets of VTK there is a new macro that simplifies object factory creation for implementation modules and two new python scripts that inspect and display module dependencies. We've replaced the old setup.py driven python installation code with a set of standard CMake installation rules and reworked the installation rules to allow the generation Java Maven packages for any platform. Starting with this release we use CMake's ExternalData module to obtain the specific versions of regression test data files that are needed. This reduces false negative results in automated testing of proposed patches to VTK. VTK's minimum required CMake version is now 2.8.8.</div>
<div><br></div><div>Starting with this release Kitware will be publishing, in addition to the usual source, data and documentation packages, runnable binaries of VTK's python shell interface for Mac, Windows and Linux operating systems. These replace and improve upon the old vtk-tcl.exe windows-only package, and like it are intended to allow users to quickly get started using VTK without the hassle of building it from source.</div>
<div><br></div><div>Besides these areas, VTK has numerous other focused improvements. Some of the most notable of these are:</div><div>* Fixes to make VTK compatible with FreeBSD, OSX 10.9, c++11 and Qt 5.</div><div>* Allow VTK to use the system installed netCDF library.</div>
<div>* C++ methods that take pointers to structs are now accessible in python.</div><div>* Several new information visualization capabilities that are summarized in the kitware blog post at <a href="http://www.kitware.com/blog/home/post/554">http://www.kitware.com/blog/home/post/554</a>.</div>
<div>* A new set of JOGL based grapical components which keep VTK compatible with Java 7 on Mac OSX.</div><div>* Improvements to text rendering such as better unicode support, font file loading and GL2PS label export.</div>
<div>* Added support for rendering double precision numbers in the charts.</div><div><br></div><div>We hope you enjoy this release of VTK! As always contact Kitware and the mailing lists for assistance.</div><div><br></div>
<div>David E DeMarle<br>Kitware, Inc.<br>R&D Engineer<br>21 Corporate Drive<br>Clifton Park, NY 12065-8662<br>Phone: 518-881-4909</div>
</div>